Well, like I said, don't put yourself out, but there is definitely value on my end at least from at least confirming it's a super rare (and maybe not consistently reproducible) bug. It would let me rest easy at night and maybe even throw together a simple tool to simple let people delete dupes or something. Mostly I just want to make sure there is a more fundamental flaw in there, which is fairly plausible given the complexity this system has reached (particularly the steps I've had to take to build it up over time, rather than from the ground-up knowing in advance all the places I would eventually decide to take it).
Anyway, I appreciate all these efforts quite a bit. I'm gonna try to trace the process once more. I think the browser thing is less likely than a few other things, anyway, so we'll see if the last patch I uploaded this afternoon stops it from happening. Obviously it's rare enough that we won't really "know" but if you're fiddling with these regularly for a month or two and it hasn't recurred, I'll call that a win.
Well, half a win, since my "fix" is less "figure out why there's two copies" and more "just delete the copies before they make their way into the database." Not as good as figuring out why it happened exactly, but acceptable.
Anyway, I appreciate all these efforts quite a bit. I'm gonna try to trace the process once more. I think the browser thing is less likely than a few other things, anyway, so we'll see if the last patch I uploaded this afternoon stops it from happening. Obviously it's rare enough that we won't really "know" but if you're fiddling with these regularly for a month or two and it hasn't recurred, I'll call that a win.
Well, half a win, since my "fix" is less "figure out why there's two copies" and more "just delete the copies before they make their way into the database." Not as good as figuring out why it happened exactly, but acceptable.